Takayo Ueno (1), Mayu Osawa (1), Tomoko Ueki (1), Yasuhiko Imai (1), Phyllis Chan (2), Tushar Garimella (2), Timothy Eley (2), Malaz AbuTarif (2), Richard Bertz (2)
(1) Bristol-Myers K.K., Research and Development, Tokyo, Japan (2) Bristol-Myers Squibb, Research and Development, NJ, USA
Objectives:The approval of the combination therapy of Daclatasvir (DCV), Hepatitis C Virus (HCV) NS5A inhibitor and Asunaprevir (ASV), HCV NS3 inhibitor in Japan, represented the world’s first approval of an interferon (IFN) and ribavirin (RBV)-free HCV treatment. The current analysis characterized the relationship between exposures of DCV and ASV and sustained virological response (SVR) in Japanese subjects who are HCV genotype (GT) 1b non-responders to pegIFNa/RBV or IFNβ/RBV, and IFN based therapy ineligible naive/intolerant receiving DUAL (DCV+ASV) and provided insight into patient covariates that are most closely associated with efficacy.
Methods:The relationship between the probability of achieving SVR at 12 weeks post treatment (SVR12) and Cavgss estimated from PPK models for DCV and ASV, was described using a logistic regression (LR) model with data from two phase 2 and 3 studies in Japanese HCV GT-1b subjects (N=265). The functional form characterization, which describes a relationship between DCV and ASV Cavgss and SVR12 and covariates identification (demographic, laboratory, prognostic and treatment covariates) were investigated during the model development steps. Non-parametric bootstrap and visual predictive check stratified by significant covariates were used for model evaluations.
Results:A linear LR model with slopes for DCV and ASV Cavgss, and interaction between DCV and ASV was identified as the base model for SVR12. In the final model, the slope for DCV Cavgss was statistically significant, but slope for ASV Cavgss and interaction between DCV and ASV were not. Among all the covariates screened, only the presence of NS5A baseline resistance mutation (BRM) Y93H was retained as a significant covariate in the final model. There was no evidence of a clinically meaningful effect of the following covariates on SVR rate: age >= 65 years, body weight, gender, creatinine clearance, ALT level, IL28B GT, baseline viral load, patient type, cirrhosis (yes/no), Study (phase 2/3), and OATP haplotype.
Conclusions:The ER model demonstrated a shallow relationship between DCV exposure and SVR12, and a flatter and non-statistically significant relationship between ASV exposure and SVR12. The presence of the NS5A BRM Y93H was a significant predictor of lower SVR12. Overall the ER model supported the high SVR12 rates for the DUAL combination in GT-1b HCV infected Japanese subjects and no dose adjustment is needed based on any of the covariates tested.
Reference: PAGE 24 (2015) Abstr 3319 [www.page-meeting.org/?abstract=3319]
Poster: Drug/Disease modeling - Infection